Plasmid-Mediated QnrS2 Determinant from a Clinical Aeromonas veronii Isolate

The main objective of this study was to determine the prevalence of the Qnr determinants in clinical and environmental Aeromonas spp. A total of 52 Aeromonas sp. isolates identified by biochemical methods (5), 25 isolated from natural waters (1) and 27 isolated from clinical samples from hospitals in Valencia, Spain, were tested for quinolone resistance by the disk diffusion method (4) (nalidixic acid, 30 μg; oxolinic acid, 2 μg; flumequine, 30 μg; ciprofloxacin, 5 μg; and levofloxacin, 5 μg). Among the studied isolates, 27 showed resistance to nalidixic acid and susceptibility to ciprofloxacin, 24 isolates were susceptible to both nalidixic acid and ciprofloxacin, and only 1, the A. veronii A272 clinical isolate, was resistant to both nalidixic acid and ciprofloxacin. The isolates resistant to nalidixic acid were also resistant to oxolinic acid and flumequine. Moreover, A. veronii A272 was the only one resistant to levofloxacin. Screening of the qnrA, qnrB, and qnrS genes was performed by multiplex PCR using a set of specific primers for all isolates. Bacterial strains positive for each qnr gene were used as positive controls (Klebsiella pneumoniae UAB1 for qnrA, Escherichia coli J53 pMG252 for qnrB, and E. coli J53 pMG298 for qnrS) and were run in each batch of tested samples. Only an A. veronii clinical isolate (A. veronii A272) presented a qnr gene, which showed 100% homology with the qnrS2 gene previously reported in an isolate from the bacterial community of a wastewater treatment plant in Germany (2) and in a non-Typhi Salmonella clinical isolate in the United States (6).

The qnrS2-carrying strain was identified as A. veronii by sequencing of the 16S rRNA gene (10). The MICs for nalidixic acid, ciprofloxacin, and norfloxacin were determined using the Etest method (AB Biodisk, Solna, Sweden). CLSI breakpoints were used to define susceptibility (4). The MICs showed by this strain were >256 mg/liter, 8 mg/liter, and 12 mg/liter to nalidixic acid, ciprofloxacin, and norfloxacin, respectively (Table 1). PCR amplification of the quinolone resistance-determining regions of the gyrA and parC genes was performed with primers previously described (7). A. veronii CECT 4260 and CECT 4258 strains, susceptible to quinolones, were included. The nucleotide and deduced protein sequences were analyzed with software available on the Internet at the National Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov). Comparison of the deduced amino acid sequences of GyrA and ParC with those of susceptible A. veronii CECT 4258 and CECT 4260 strains showed that A. veronii A272 carried mutations in the quinolone resistance-determining regions of both the gyrA and parC genes (Table 1). As previously reported, the same mutations were found in a strain that showed a MIC to ciprofloxacin of 0.5 mg/liter (7). Therefore, the level of resistance to these antibacterial agents could be explained by these two mutations in the gyrA and parC genes plus the additive effect of the presence of the QnrS2 determinant as previously suggested by Martinez-Martinez et al. (9).

TABLE 1.

Quinolone susceptibilities of the A. veronii A272 clinical isolate, the E. coli J53 isolate, and the E. coli J53 isolate transformed with plasmid pA272

Strain	Amino acid change		QnrS2a	MIC to quinolones (mg/liter)b
	GyrA	ParC		NAL	CIP	NOR
A. veronii CECT 4258	Ser83	Ser80	−	<0.25	<0.02	<0.02
A. veronii CECT 4260	Ser83	Ser80	−	<0.25	<0.02	<0.02
A. veronii A272	Ile83	Ile80	+	>256	8	12
E. coli J53			−	4	0.023	0.19
E. coli J53 + pA272			+	>256	0.5	2

^a−, negative; +, positive.

^bNAL, nalidixic acid; CIP, ciprofloxacin; NOR, norfloxacin.

Conjugation experiments were performed by the liquid mating-out assay using rifampin-resistant E. coli J53 and an environmental Aeromonas subsp. resistant to rifampin as recipient strains and nalidixic acid as the selective agent. The conjugation experiments provided negative results. After the extraction and purification of the qnrS2-containing plasmid (pA272) (the qnrS2-containing plasmid showed a size ranging from 48.5 to 97.0 kb [Lambda ladder PFG marker; New England Biolabs, Ipswich, MA]), a transformation experiment was done with rifampin-resistant E. coli J53 by electroporation. The quinolone MICs of the recipient E. coli J53 strain and of the transformed strain, determined by Etest, revealed a 10- to more than 64-fold increase in the transformant (Table 1). In addition, we did not observe increased resistance to β-lactam antibiotics, aminoglycosides, chloramphenicol, and tetracycline in the transformed E. coli strain (data not shown).

To analyze the genetic context of the qnrS2 gene, the DNA of A. veronii A272 was digested with MspI “C*CGG” (recognition site) (New England Biolabs, Ipswich, MA), which does not have recognition sites in the qnrS2 gene. The fragments obtained were autoligated overnight at 16°C using T4 DNA ligase (Promega Biotech Ibérica, Madrid, Spain) and used as a template for a PCR with inverse primers designed from the qnrS2 gene sequence (qnrs2invF, 5′-GAACAGCTTCTCGAAGCGTTG-3′, and qnrs2invR, 5′-ACTGTGGTGTCGATATGTGTG-3′). The resulting bands were sequenced using the BigDye Terminator v3.1 cycle sequencing kit (Applied Biosystems, Warrington, United Kingdom). The sequence obtained around the qnrS2 gene (approximately 300 bp from each side) showed that this gene was inserted into an mpR gene as previously reported by Cattoir et al. (3).

This is the first time that a qnrS-containing plasmid in an Aeromonas sp. clinical isolate has been described. Up to now, qnr determinants have only been reported in Enterobacteriaceae (8, 11-15), except for a recent report on a qnrS-containing plasmid found in environmental A. caviae (A. punctata) and A. media isolates (3). The identification of a qnrS2 gene in a clinical isolate not within the Enterobacteriaceae family emphasizes the versatility of these determinants to spread among the different bacterial species with the consequent potential risk for human health.